Extraction of Domain-Specific Relations in Spanish Labour Law

The code and data of the related project.

Structure

• data contains the raw and the manually annotated data
• model_training contains the scripts to train the model

OLD Training with R-BERT

We use the cool and robust R-BERT model to recognize relations in the text.

How to run

Check train.conf to provide the correct config values.

0. Create virtual environment (see this link) and install requirements (pip3 install -r requirements.txt).
1. Convert the manually annotated data

   cd model_training
   python3 convert.py --config_path train.conf

2. Train model

   cd model_training
   python3 train.py --config_path train.conf

More recent results

We use a modification of GRIT model, our code is not published yet, but you can find the produced predictions at data/predictions/grit/estatuto_predicted.json.

Publications

Recently our paper was accepted to DeepOntoNLP @ ESWC 2021!

To cite that paper:

@inproceedings{DBLP:conf/esws/Martin-ChozasR21,
  author    = {Patricia Mart{\'{\i}}n{-}Chozas and
               Artem Revenko},
  editor    = {Sarra Ben Abb{\`{e}}s and
               Rim Hantach and
               Philippe Calvez and
               Davide Buscaldi and
               Danilo Dess{\`{\i}} and
               Mauro Dragoni and
               Diego Reforgiato Recupero and
               Harald Sack},
  title     = {Thesaurus Enhanced Extraction of Hohfeld's Relations from Spanish
               Labour Law},
  booktitle = {Joint Proceedings of the 2nd International Workshop on Deep Learning
               meets Ontologies and Natural Language Processing (DeepOntoNLP 2021)
               {\&} 6th International Workshop on Explainable Sentiment Mining
               and Emotion Detection {(X-SENTIMENT} 2021) co-located with co-located
               with 18th Extended Semantic Web Conference 2021, Hersonissos, Greece,
               June 6th - 7th, 2021 (moved online)},
  series    = {{CEUR} Workshop Proceedings},
  volume    = {2918},
  pages     = {30--38},
  publisher = {CEUR-WS.org},
  year      = {2021},
  url       = {http://ceur-ws.org/Vol-2918/paper4.pdf},
  timestamp = {Tue, 10 Aug 2021 16:26:49 +0200},
  biburl    = {https://dblp.org/rec/conf/esws/Martin-ChozasR21.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}